Efficient deep neural network for photo-realistic image super-resolution

  title={Efficient deep neural network for photo-realistic image super-resolution},
  author={Namhyuk Ahn and Byungkon Kang and Kyung-ah Sohn},
  journal={Pattern Recognit.},

Exploiting Distortion Information for Multi-degraded Image Restoration

This work introduces a distortion information-guided restoration network, which exploits the conditional distortion information when reconstructing a given image, and proposes a new dataset that is a holistic multi-distortion dataset.

Single-Image Super-Resolution Neural Network via Hybrid Multi-Scale Features

A novel multi-scale super-resolution neural network (HMSF), which is more lightweight, has fewer parameters, and requires less execution time, but has better performance than the state-of-the-art methods.

Lightweight image super-resolution with enhanced CNN

Lightweight Modules for Efficient Deep Learning Based Image Restoration

This paper proposes several lightweight low-level modules which can be used to create a computationally low cost variant of a given baseline model, and extensively validate the findings on three popular tasks of image inpainting, denoising and super-resolution.

CVEGAN: A Perceptually-inspired GAN for Compressed Video Enhancement

The proposed CVEGAN generator benefits from the use of a novel Mul2Res block (with multiple levels of residual learning branches), an enhanced residual non-local block (ERNB) and an enhanced convolutional block attention module (ECBAM) to improve the representational capability.

Enhancing the Quality of Cellular Camera Video With Convolutional Neural Network

The CNN (Convolutional Neural Network) method can improve the image of video recordings that have poor quality.

MFRNet: A New CNN Architecture for Post-Processing and In-loop Filtering

A novel convolutional neural network architecture, MFRNet, for post-processing (PP) and in-loop filtering (ILF) in the context of video compression with significant and consistent coding gains over both anchor codecs and also over other existing CNN-based PP/ILF approaches based on Bjøntegaard Delta measurements.

Super-Resolution Appearance Transfer for 4D Human Performances

A pipeline is proposed for super-resolution appearance transfer from high-resolution static capture to dynamic video performance capture to produce super- Resolution dynamic textures, which addresses two key problems: colour mapping between different camera systems; and dynamic texture map super- resolution using a learnt model.



Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

SRGAN, a generative adversarial network (GAN) for image super-resolution (SR), is presented, to its knowledge, the first framework capable of inferring photo-realistic natural images for 4x upscaling factors and a perceptual loss function which consists of an adversarial loss and a content loss.

A Fully Progressive Approach to Single-Image Super-Resolution

To obtain more photorealistic results, a generative adversarial network (GAN) is designed, named ProGanSR, that follows the same progressive multi-scale design principle and constitutes a principled multi- scale approach that increases the reconstruction quality for all upsampling factors simultaneously.

Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution

This paper proposes the Laplacian Pyramid Super-Resolution Network (LapSRN) to progressively reconstruct the sub-band residuals of high-resolution images and generates multi-scale predictions in one feed-forward pass through the progressive reconstruction, thereby facilitates resource-aware applications.

EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesis

This work proposes a novel application of automated texture synthesis in combination with a perceptual loss focusing on creating realistic textures rather than optimizing for a pixelaccurate reproduction of ground truth images during training to achieve a significant boost in image quality at high magnification ratios.

Fast and Accurate Image Super-Resolution with Deep Laplacian Pyramid Networks

This paper proposes the deep Laplacian Pyramid Super-Resolution Network for fast and accurate image super-resolution, and utilizes the recursive layers to share parameters across as well as within pyramid levels, and thus drastically reduce the number of parameters.

ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks

This work thoroughly study three key components of SRGAN – network architecture, adversarial loss and perceptual loss, and improves each of them to derive an Enhanced SRGAN (ESRGAN), which achieves consistently better visual quality with more realistic and natural textures than SRGAN.

Image Super-Resolution Using Dense Skip Connections

A novel single-image super-resolution method is presented by introducing dense skip connections in a very deep network, providing an effective way to combine the low-level features and high- level features to boost the reconstruction performance.

Accelerating the Super-Resolution Convolutional Neural Network

This paper aims at accelerating the current SRCNN, and proposes a compact hourglass-shape CNN structure for faster and better SR, and presents the parameter settings that can achieve real-time performance on a generic CPU while still maintaining good performance.

Image Super-Resolution via Deep Recursive Residual Network

This paper proposes a very deep CNN model (up to 52 convolutional layers) named Deep Recursive Residual Network (DRRN) that strives for deep yet concise networks, and recursive learning is used to control the model parameters while increasing the depth.